Jim,
Thanks for the link. You have to be very careful believing anything you read out there in cyberspace. Quite often the abstracts will fool you. That particular study was done in Sweden on Ta and T1 bladder cancer patients treated between 1963 and 1972. It was done pre-BCG and reflects the natural course of the disease without BCG treatments.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=7752327&dopt=Abstract
Many of the studies I have found don't have a large enough base group to be statistically significant, or they don't differentiate between the specifics of the patients. Sometimes it isn't until you dig down into the report (not the abstract) that you find that only half the patients were given BCG or all the patients entered into the study were actually entered on their first high grade recurrance (instead of their primary recurrance)... Issues like that stack the deck and skew the statistics.
Other studies may include patients catorigized as stage T1, but because no muscle was removed during their initial TURB, their resection was never actually complete. So, they may actually have had a T2 or T3 invasive tumor and even with BCG they progress/die - such a study may incorrectly report having a larger progression rate for T1 disease than other studies.
Other studies may be age biased. I'm sure there are many senile 98 year olds in nursing homes, who if followed up regularly with top care would not progress or die of invasive bladder cancer. However some studies just take a retroactive look at death certificates and that senile 98 year old becomes a disease specific progression/death statistic on some 20 year study. How does that statistic compare to a 35 year old going to quarterly cystos and having BCG treatments? It doesn't.
The EORTC risk tables are the best study I have seen to date to help guage an individuals specific recurrance and progression rate based on his/her tumor characteristics.